Semantic Integration of Goal and Business Process Modeling

Semantic Integration of Goal and

Business Process Modeling

Evellin C.S. Cardoso, Paulo Sérgio dos Santos Jr., João Paulo A. Almeida,

Renata S.S. Guizzardi, Giancarlo Guizzardi

e-mail: {ecardoso, paulossjunior, jpalmeida, rguizzardi, gguizzardi}@inf.ufes.br

Ontology and Conceptual Modeling Research Group (NEMO) Computer Science Department, Federal University of Espírito Santo (UFES)

Av. Fernando Ferrari, s/n, Vitória, ES, Brazil

Abstract. While business processes and business goals are considered intrinsically interdependent, a comprehensive modeling approach that includes

both the business process and the goal perspectives is still lacking. This paper proposes a semantic integration between the domains of goal modeling and business process modeling. We integrate the ARIS framework with the Tropos goal modeling language. While ARIS is widely employed for business process modeling, it offers an overly simplistic set of goal-related concepts. In contrast, Tropos offers a rich set of goal-related concepts (and associated goal analysis methods), while refraining from addressing business process modeling in detail. In order to investigate the relation between the Tropos modeling constructs and the ARIS elements, we propose an ontological account for both architectural

domains through the usage of the UFO ontology.

Keywords: Semantic Integration, Goal Modeling, Business Process Modeling EPC, TROPOS, ARIS

l. INTRODUCTION

The increasing competitiveness drives organizations to promote change in an

attempt to improve the quality of the services and products they offer. In recent years,

many of the efforts related to managing change in organizations have been conducted

in the scope of Business Process Reengineering (BPR) activities [8][20][21][22]. BPR

is based on the assumption that changes in business processes should generate radical

improvements in critical performance measures (such as cost, quality, service and

speed)[20].

However, predicting how a given enterprise environment should respond to changes by simply adopting a business-process centered view is unfeasible since there

are a large number of issues to be considered, such as infrastructure, power and

politics, managerial control, organizational culture, among others [35]. Given this

multitude of issues, understanding an organizational setting often requires a number

of perspectives [35].

2 Evellin C.S. Cardoso, Paulo Sérgio dos Santos Jr., João Paulo A. Almeida,


Among these perspectives, the domain of “motivation” has been recognized as an

important element of enterprise architectures [36] as highlighted in Zachman

framework’s motivation column [37]. Goal modeling is the artifact employed for

capturing the motivational aspect and strategies behind the organizational practices

[35], helping in clarifying interests and intentions from different stakeholders [19].

Moreover, by adopting goal modeling, the organizations can systematically express

the choices behind multiple alternatives and explore new possible configurations for an organizational setting. This is essential for business improvement since changes in

a company’s strategy and business goals have significant consequences within all

domains of the enterprise [23][24].

Since business process and goals are intrinsically interdependent, establishing an

alignment between both domains arises as a natural approach. The central idea is to

create enterprise models that describe not only the entities in a business context, but

also include motivations for those entities [2].

This paper contributes to this vision by proposing a semantic integration between

the domains of goal modeling and business process modeling. We integrate the ARIS

framework [32] with the Tropos methodology and modeling language [4][5]. While

ARIS is widely employed for business process modeling, it offers an overly simplistic set of goal-related concepts. In contrast, Tropos offers a rich set of goal-related

concepts (and associated goal analysis methods), while refraining from addressing

business process modeling in detail.

Since each modeling language focuses on different architectural domains of the

organization (which is manifested through the existence of different sets of concepts

in each modeling language), we use an ontological approach for bridging the semantic

gap between the two modeling languages. This involves the interpretation of the

related concepts in each of these languages and a subsequent harmonization of the

languages. For this interpretation, we employ a foundational ontology, i.e., a formal

and ontologically sound system of domain-independent categories. In particular, we

make use here of the Unified Foundational Ontology (UFO) [16] as our semantic

foundation. This paper is further structured as follows: section 2 presents the relevant

fragments of the ARIS and Tropos metamodels; section 3 discusses the interpretation

of these metamodels in terms of UFO and presents the integration of both approaches;

section 4 presents our conclusions and identifies topics for further investigation.

2 THE ARIS AND TROPOS METAMODELS

Before interpretation, we must identify the relevant language constructs and their

relations. This is discussed in this section, which presents fragments of the ARIS and

Tropos language metamodels (represented here in Ecore [11]).


Business Process Modeling 3

2.1 The Tropos Metamodel

The i* framework [35], consists in an agent-oriented conceptual framework whose

focus is on intentional characteristics of organizational actors. The Tropos

methodology has been conceived with basis on the i* framework and adopts the same

concepts in early requirements stages for software development [34]. Since Tropos captures the intentions and motivations of organizational actors, we use the Tropos

methodology for goal modeling, adopting a subset of the Tropos modeling language.

The language is structured in terms of two main components: the Actor Diagram

and the Goal Diagram. The former describes the organizational context in terms of

dependency relationships between actors, while the latter describes the actors’ goals

and rationales in order to justify the actors’ relationships and their adoption of

particular plans. The metamodels of these two diagrams are extensively described in

[12]. In this section, we present fragments of these metamodels and provide some

examples of usage of the modeling constructs.

Figure 1. Metamodel of the Actor Diagram [19].

Figure 1 depicts the metamodel of the Actor Diagram. In this metamodel, Actor is

the agent-oriented concept which represents an intentional entity of the organizational

setting. An Actor is specialized into other three concepts, namely: Agent, Role and

Position. A Role is a characterization of properties that apply to actors playing that

role in a given social domain (it is transferable to other individuals). An Agent is an

actor which displays a physical existence, such as human individuals, hardware or software agents. Finally, a Position comprises in a set of roles which is performed by

an agent [35]. We say that an agent occupies zero or more positions and plays zero or

more roles. Further, a position is said to cover one or more roles [5][12][35].

With respect to goals, Tropos relies on two primitives for goal modeling: hardgoals

and softgoals. The language has a general concept Goal which, in its turn is refined

into these concepts. A Goal, according to [35], is defined as a condition or state of

affairs in the world that the actor would like to achieve.

Hardgoals are defined as goals whose satisfaction can be objectively defined [5].

Conversely, softgoals are “subject to interpretation” [35], “imprecise, subjective,

context-specific, and ideal” [25] and therefore have no objective satisfaction criteria.

This different nature of achievement is denoted in the terms used for stating goal fulfillment: it is said that hardgoals are satisfied while softgoals are satisficed [27].



The actor diagram also identifies plans which are executed by agents. The

relationship between plans and goals rests on the fact that goals represent “a set of

states of affairs (i.e. a set of world states)”, while plans “represent, at an abstract level,

a way of doing something. The execution of plan can be a means for satisfying a goal

or for satisficing a softgoal” [5].

Commonly, the actors cannot satisfy their goals in isolation and, as consequence,

they engage in dependency relations with other actors. These are relations are also represented in actor diagrams. A dependency represents an agreement between two

actors where one actor (the depender) depends on another (the dependee) to fulfill a

goal, perform a plan or deliver a resource (the dependum) [12]. Resources [35] are

intentional objects (usually obtained as a finished product from a deliberation

process).

Figure 2. Metamodel of the Goal Diagram [12]

In the Goal Diagram (whose metamodel is depicted in Figure 2), the central

concept of goal is represented by the Goal metaclass [12]. Goals can be analyzed, from the point of view of an Actor, by three types of relationships among them:

means-ends links, AND/OR decomposition and contribution links. Means-end links

aim at capturing which plans and resources provide means for achieving a goal

(therefore, a means-ends link is a ternary relationship between an Actor, a Goal (the

end) and a Plan or Resource (the means) [12]). Further, there are two types of

decompositions (specified via an attribute of the metaclass Boolean Decomposition):

AND-decomposition and OR-decomposition. An AND decomposition supports a goal

to be decomposed in a series of sub-goals; while an OR decomposition allows

modeling alternative ways of achieving a goal. Contribution links identify goals that

can contribute positively or negatively in the attainment of the goal to be analyzed

(thus, it is a ternary relation between An Actor and two goals).



2.2 Metamodels for the ARIS Method

Figure 3 shows a fragment of the metamodel of the business process modeling and

goal modeling languages used in the ARIS Method. This fragment was excavated in

our earlier work by using the approach described in [30] and defines the abstract

syntax of the language as currently supported by the ARIS Toolset. The main metaclasses for business process modeling in this fragment are: Participant,

Objective, Event, Rule and Function. Business processes are modeled in diagrams

know as Event-driven Process Chains (EPC). The main metaclasses for objective

modeling in this fragment are: Objective, Critical Factor, Product/service and

Function. These metaclasses are used in an Objective Diagram.

Figure 3. Fragment of the metamodel for Business Processes and Goals in ARIS

The Participant abstract metaclass subsumes the following metaclasses:

Organization Unit Type, Organization Unit, Position, Person Type, Person, Group

and Employee Variable. These organization elements belong to the ARIS

Organizational diagram and are referenced in an EPC to describe participants in

organizational activities. According to [32], the Organization Unit metaclass represents an entity that is responsible for achieving organizational goals. The

Position metaclass represents the smallest organizational unit possible (a particular

job position). The Person metaclass is used to represent a person who is assigned to

organization. The Person Type metaclass represents a role performed by one or more

persons, positions, groups or organizational units a business process [31][32]. The

Group represents a group of employees (Person) or a group of organizational unit

(Organizational Unit) that work together to achieve a goal.

The Function metaclass is a basic element for EPC process modeling. According to

the ARIS documentation, the element Function represents either a technical task or a

task performed on some object, with the purpose of achieving one or more business

goals [32]. A task can be performed by either a person or an application system [32],

and has inputs – such as information or raw material – and outputs, such as new information or products. Furthermore, tasks can consume and create organizational

resources during their execution [1]. The carries out meta-association between the



Participant and Function elements indicates that one or more participants of the

business process will be responsible for performing the task. Due to space constraints

we refer the reader to [31] for a full treatment of the Event and Rule metaclasses as

well as the is predecessor of, activates, creates meta-associations. For the purposes of

this paper, it is sufficient to assume that these modeling elements enable different

types of specification of behavior.

An objective diagram “models a hierarchy of business objectives along with their critical success factors and the Functions and Products that support achievement of

these objectives” [9]. According to the ARIS documentation, “a product/service is

performed in the course of a value-added process. It is the result of a human act or a

technical process. A product/service can represent either a service or a product”

[10][32]. The Critical Factor metaclass represents the aspects which need to be

considered in order to reach a particular objective [32] (and follows the Critical

Success Factor definition by Rockart [29]).

The language has opted for modeling the relationship between goals (represented

by the Objective metaclass) and Functions since the execution of functions can be

seen as operations applied to objects for the purpose of supporting one or more goals

[32]. This relationship is denominated as “supports of” relationship. Goals (Objectives) and their relationships are also modeled in this view. Goals can be linked

with one another with a subordinate goal supporting several overriding goals (through

the “belongs to” relationship).

3 INTERPRETATION OF ARIS AND TROPOS

3.1 Ontological Foundations

In the sequel, we discuss a fragment of UFO in line with the purposes of this

article. For a full discussion regarding this foundational ontology, one should refer to

[16][ 14].

We start with the fundamental distinction between universals and individuals. The

notion of universal underlies the most basic and widespread constructs in conceptual

modeling. Universals are predicative terms that can possibly be applied to a multitude

of individuals, capturing the general aspects of such individuals. Individuals are entities

that exist instantiating a number of universals and possessing a unique identity.

Further, UFO makes a distinction between the concepts of Endurants and Events

(also known as Perdurants). Endurants are individuals said to be wholly present

whenever they are present, i.e., they are in time, in the sense that if we say that in circumstance c1 an endurant e has a property P1 and in circumstance c2 the property

P2 (possibly incompatible with P1), it is the very same endurant e that we refer to in

each of these situations. Examples of endurants are a house, a person, the moon, a

hole, an amount of sand. Events (Perdurants), in contrast, are individuals composed by

temporal parts, they happen in time in the sense that they extend in time accumulating

temporal parts. An example of an Event is a business process. Whenever an Event

occurs, it is not the case that all of its temporal parts also occur. For instance, if we



consider a business process “Buy a product” at different time instants when it occurs,

at each of these time instants only some of its temporal parts are occurring.

A Substantial is an Endurant that does not depend existentially on other Endurants,

roughly corresponding to what is referred by the common sense term “Object”. In

contrast with Substantials, we have Moments (also known as particularized properties,

objectified properties and Tropes). Moments are existentially dependent entities, i.e.,

for a Moment x to exist, another individual must exist, named is bearer. Examples of Substantials include a person, a house, a planet, and the Rolling Stones; examples of

Moments include the electric charge in a conductor, a marriage, a covalent bond as

well as mental states such as individual Beliefs, Desires and Intentions (or internal

commitments). The last three examples fall in the subcategory of Mental Moments.

UFO also adds distinctions concerning the intentionality of events to this basic

core. Examples include the concepts of Action, Action Universal, Action Contribution and

Agent.

Actions are intentional events, i.e., events which instantiate a Plan (Action Universal)

with the specific purpose of satisfying (the propositional content of) some Intention of

an Agent. The propositional content of a commitment is termed a Goal. Only agents

(entities capable of bearing intentional moments) can perform Actions. As events, actions can be atomic (Atomic Action) or complex (Complex Action). While an Atomic Action is

an action event that is not composed by other action events, a Complex Action is a

composition of at least two basic actions or Participations (that can themselves be

atomic or complex).

Participations can themselves be intentional (i.e., Actions) or non-intentional Events.

For example, the stabbing of Caesar by Brutus includes the intentional participation of

Brutus and the non-intentional participation of the knife. In other words, we take that

it is not the case that any participation of an agent is considered an action, but only

those intentional participations called Action Contributions.

Figure 4. Fragment of UFO.

The category of agents further specializes in Physical Agents (e.g., a person) and

Social Agents (e.g., an organization, a society). In an analogous manner, objects can also be categorized as Physical Objects (e.g., cars, rocks and threes) or Social Objects



(e.g., a currency, a language, the Brazilian constitution). Agents can also be further

specialized into Human Agent, Artificial Agent and Institutional Agent, which can be

represented, respectively, by human beings, computationally-based agents and

organization or organizational unit (departments, areas and divisions). Institutional

Agents are composed by a number of other agents, which can themselves be Human

Agents, Artificial Agents or other Institutional Agents.

Each Institutional Agent has a Normative Description associated to it. Moreover, this Institutional Agent defines a context in which a normative description is recognized (see

relationship recognized by in Figure 5 ). We can state then that Normative Descriptions

are social objects that create social entities recognized in that context. Examples

include Social Roles (e.g., president, manager, sales representative), Social Role Mixins

(whose instances are played by entities of different kinds, e.g., customer, which can

be played by persons and organizations), Social Agent Universals (e.g., a political party,

an education institution), Social Agents (e.g., the Brazilian Labour Party, the University

of Twente), Social Object Universals and other Social Objects (e.g., a piece of legislation,

a currency) or other Normative Descriptions [1]. A Normative Description that defines

social individuals in the context of an institutional agent is termed a Constitutive

Normative Description [1].

Figure 5. Fragment of UFO with social aspect.

3.2 Ontological Analysis of the ARIS Metamodels

According to [17] and [33], a business process can be defined as a collection of

interrelated organizational tasks, initiated in a response to an event, which aim at

achieving one or more organizational goals. In other words, a business process

describes a type of organizational task that must be performed to achieve one or more

organizational goals. Since EPCs are used for business process modeling, we can say

that, collectively, the elements of an EPC diagram can be interpreted as a Complex

Action Universal of UFO [14]. According to [15][32], a Function can be defined in

several abstraction and refinement levels. Therefore, in [31] the Function element was

interpreted as Action Universal.

Since organizational units can be decomposed recursively into smaller

organizational units [32], we interpret the Organizational Unit metaclass representing

a particular kind of substantial, namely, an Institutional Agent. Similarly, the



Organization Unit Type metaclass has been interpreted as an Institutional Agent

Universal. The Position metaclass does not represent an organization unit, but a Social

Role instantiated by a Human Agent (Person) [1].

Person Type is an element to represent a role which can be instantiated by different

entities (persons, and, despite the name suggesting the contrary, organizational units).

Thus, the Person Type element has been interpreted as a Social Role Mixin. It is often

used in the scope of a business process to avoid tying specific agents to business processes, differently from Position which is a social role defined in the

organizational structure.

The Group metaclass can be interpreted as a Collective Social Agent or as an

Institutional Agent. The first interpretation occurs when Group represents a collection of

agents playing the same role. The second interpretation occurs when Group

represents a collection of agents each of which with a different role [16].

The Objective element is used to represent a business objective associated with a

business process (Function) or business product/service (Product/Service). While the

element’s name would suggest a correspondence to the Goal concept in UFO, this

interpretation is far from trivial. This is because UFO Goals are necessarily associated

with a particular Agent (they are the propositional content of an agent’s intention). A viable interpretation is that the (Institutional) Agent which has the Goal is the owner

of the business process (Function) which supports the Objective such as, for instance,

the Organization (Unit) which (partial) behavior is described by that process

universals or, alternatively, a Social Role within an organizational structure which

contains that behavior specification as part of its definition. This notion of “owner”,

however, is not directly modeled in ARIS, although it is implied by Scheer [32] when

referring to “corporate goals”, which are necessarily present at an organization

whenever a business process exists.

The belongs to relation between Objectives defies a precise definition, since it may

refer to a number of different relations, not distinguishing conjunctions or

disjunctions of propositions. Further, there is very little explanation in the ARIS

literature concerning the role of Product/Service in an ARIS objective diagram. Thus, we will refrain from providing a complete interpretation here; instead of adopting the

relations in the ARIS Object Diagram metamodel, we will use the richer relations

between Tropos Goals, Agents, Plans and Resources as discussed in the next section.

3.3 Ontological Analysis of the TROPOS Metamodels

We start our ontological analysis with the fragment which considers the specializations of the metaclass Actor (the interpretation of this part of the metamodel

is based on the interpretation discussed in [18]).

We interpret the metaclasses Agent and Role in Tropos as the concepts of Agent and

Social Role in UFO (respectively). The metaclass Position is also interpreted as a Social

Role. This interpretation is supported by the fact that Tropos positions are defined

solely with the purpose of aggregating different roles. Since an agent role is defined

by the set of social moment universals (commitments and claims implied by the role),

an aggregation of roles is also a role in itself, i.e., a universal capturing a set of social

moment universals [18]. The abstract metaclass Actor is introduced to capture general



relations between Agent, Roles, Positions and other modeling elements and, thus, it

has no specific interpretation in itself.

We interpret Tropos goals as Goals in UFO. Goals in UFO are sets of intended

states of affairs of an agent. The relation between an Actor in Tropos and a Goal

(through the meta-association wants) is interpreted indirectly by making use of the

concept of Intention (or Internal Commitment) in UFO, which is a Mental Moment of an

Agent. As previously discussed, UFO contemplates a relation between Situations and Goals such that a Situation (or possibly a number of Situations) may satisfy a Goal. In

other words, since a Goal is a proposition (the propositional content of an Intention), we

have that a particular state of affairs can be the truthmaker of that proposition. This

interpretation choice seems to model directly the intention behind the concept of

hardgoal in Tropos. For the case of softgoals, a different analysis must be conducted.

The concept of softgoals does not have a uniform treatment in the Tropos

community. Sometimes, softgoals are taken to represent non-functional requirements

[7]. In other times, a softgoal is considered as a fuzzy proposition, i.e., one which can

be partially satisfied (or satisfied to a certain degree, or yet, satisficed) by Situations

[13]. We here take a different stance, namely, that a softgoal is one “subjective to

interpretation” and “context-specific”. As a consequence of this conception, for the case of softgoals, it seems to be

impossible to eliminate a judging agent (collective or individual) from the loop. Thus,

instead of considering in the ontology a new satisfices relation between Situation and

Goal which perhaps should contemplate a fuzzy threshold of satisfaction, we take a

different approach. We consider the relation of satisfaction as a ternary relation that

can hold between an agent, a goal and situation. An instance of this relation is derived

from the belief of an agent that a particular situation satisfies the goal at hand. Now,

in this view, different agents can have different beliefs about which sets of situations

satisfy a given goal. In fact, it is exactly this criterion which seems to capture the

aforementioned notion of softgoals and its differentiae w.r.t. hardgoals: (i) a goal G is

said to be a hardgoal iff the set of situations that satisfy that goal is necessarily shared

by all rational agents; (ii) a goal G is said to be a softgoal iff it is possible that two rational agents X and Y differ in their beliefs to which situations satisfy that goal.

The mapping of the Plan concept from Tropos to some UFO concept is established

in a direct manner. In section 2.1, we stated that a Plan in Tropos is a specific way of

doing something to satisfy some Goal (or satisfacing some Softgoal). From the UFO

ontology (section 3.1), we have that an Action (instance of an Action Universal) is an

intentional event performed by agents with the purpose of achieving goals.

Consequently, the Tropos Plan construct can be interpreted as an Action Universal.

The metamodel includes a relation of means-end between a Plan and a Goal. We

call attention to the point of view relation in the metamodel of Figure 2. As one can

observe, in the Tropos metamodel, the means-end relation is a ternary relation

indexed to an Agent’s (subjective) point of view. The form of this relation in the metamodel seem to corroborate our interpretation of goals just discussed. Thus, in

general, the means-end relation between a Plan and a Goal can be interpreted in the

following manner: a Plan P is a means-end to a Goal G in the point of view of Agent

A iff one or more executions of that Plan produce a post-situation which A believes to

satisfy G. The concept of Resource has been interpreted as a resource in UFO, i.e., as



a Non-agentive Substantial (or Object) which participates in a Complex Action. The relation

of means-end can also be defined between a Resource type, a Plan and an Agent, or

between a Resource type, a Goal and an Agent. The former mode of this relation can

be interpreted as follows: a Resource type R is a means-end to a Goal G in the point

of view of Agent A iff every Action which satisfies that Goal (according to A) has as

part a participation of a resource of that type. In contrast, the means-end relation

between Resource type and Plan should be interpreted as: a Resource type is a means-end to a Plan iff every Action instance of that Plan has as part a participation of a

resource of that type. Now, notice that the latter rendering of relation is actually

Agent-independent! If a Plan is taken to be an Action universal, this relation reflects

the structure of Plan and not the belief of a particular agent regarding the structure of

a Plan. In (apparent) opposition to this idea, one could argue that a Plan should not

then be interpreted as an Action universal but as an Intention to execute a particular

Action universal. Even if this view is taken, the correct alternative interpretation

would be that a Plan is an intention to instantiate a particular specialization of that

Action Universal in which resources of that type are essential participants. Still, this

would only refine the reference to a particular subtype of that Action universal. The

participation of that resource in instances of that (now more specific) Action universal would still reflect the structure of those actions, not an Agent’s subjective point of

view.

In Tropos, goals can be further structured by using different types of relations,

namely, AND-decomposition and OR-decomposition. Since Goals are taken here to

be propositions, if we have that goals G1…Gn AND-decompose goal G0, this relation

should be interpreted as: (G0 ↔ (G1 G2 …Gn)). In an analogous manner, and OR-

decomposition G1…Gn of goal G0 should be interpreted as: (G0 ↔ (G1 G2 … Gn)). Here once more, these relations reflect logical relations between propositions

and, accordingly, are independent of an Agent’s point of view (contra Figure.2).

Finally, Tropos includes a relation of contribution that can be defined between a

hardgoal and a softgoal, or between a plan and a softgoal. The idea is that a hard-goal

or plan can positively or negatively “contribute to” a soft-goal. Since soft-goals

involve subjective judgments of agents, the relation of contribution must be agent-

indexed. Thus, one should not state that G’ contributes positively to G” but that Agent

X believes that G’ contributes positively to G. One should notice that the Tropos metamodel (Fig.2) takes this relation as a ternary one indexed to an Agent’s point of

view. Further, the contribution relation can be used between a Resource and a Goal, in

the sense that the Resource is a means to a plan that the Agent believes that

contributes positively to the Goal. A fuller interpretation of this relation requires an

elaboration of the propositional content of beliefs, which is outside the scope of this

paper.

Having clarified the semantics of the modeling constructs through interpretation in

terms of UFO, we establish the correspondence between the constructs in each of the

identified fragments of Tropos and ARIS in Table 1.



Table 1. Interpretation ARIS and TROPOS.

Tropos Ontological concept (from UFO) ARIS

Agent

Agent Person

Institutional Agent Organization Unit and Group

Collective Social Agent Group

Role or Position Social Role Position and Person Type

Goal Goal

Objectives

HardGoal A Goal such that the set of situations that satisfy

that goal is necessarily shared by all rational agents;

SoftGoal A Goal such that Agents can differ in their beliefs

to which situations satisfy it.

Plan Action Universal Function

Resource Non-Agentive Substantial Product/Service

Dependency1 Dependency or

Delegation

N/A

Means-ends

(Plan and Goal)

One or more executions of that Plan produce a post-

situation that satisfies the Goal. Included in the scope of

supports relation between

Function and Objective

Means-ends

(Resource and

Goals)

Every Action which satisfies that Goal has as part a

participation of a Resource of that type

included in the scope of

supports relation between

Product/Service and Objective

Means-ends

(Resource and

Plan)

Every Action instance of the Plan has as part a

participation of a Resource of that type.

included in the scope of

input for and is consumed by

relationship between

Product/Service and Function

Negative

Contribution

(Plan, SoftGoal)

Agent believes that an execution of the Plan contributes negatively to Goal

N/A

Positive

Contribution

(Plan, SoftGoal)

Agent believes that an execution of the Plan contributes positively to Goal

Included in the scope of supports relation between

Function and Objective

Negative

Contribution

(Resource,

SoftGoal)

The Resource is a means to a Plan that the Agent believes contributes negatively to Goal

N/A

Positive

Contribution

(Resource,

SoftGoal)

The Resource is a means to a Plan that the Agent

believes contributes positively to Goal

Included in the scope of supports relation between

Product/Service and Objective

Negative

Contribution

(Goal G1,

SoftGoal G2)

Agent believes that the Goal G1 contributes

negatively to Goal G2

N/A

Positive

Contribution

(Goal G1,

SoftGoal G2)

Agent believes that the Goal G1 contributes

positively to Goal G2

Included in the scope of

belongs relation between

Objectives AND

decomposition

The propositional content of the composed Goal is

the conjunction of the component Goals

OR

decomposition

The propositional content of the composed Goals is

the disjunction of the component Goals

1 An ontological analysis of the relation of Dependency in Tropos was performed elsewhere [40] and we

refrain from this discussion here due to space constraints.



4 CONCLUSIONS AND FUTURE WORK

Although Zachman’s framework has recognized the importance of the goal domain

in its “motivation” column, Zachman did not define basic concepts for this column,

justifying that “there is a scarcity of good examples in the people, time, and

motivation columns” and stating that “the why column would be comprised of the

descriptive representations that depict the motivation of the enterprise, and the basic columnar model would likely be ends-means-ends, where ends are objectives (or

goals) and means are strategies (or methods)” [34]. So far (18 years later), no

comprehensive enterprise modeling approach has addressed the why column with

significant expressiveness while relating it to the other columns with strong semantic

underpinnings.

This work has contributed towards filling this gap by proposing a semantic

integration between the Tropos goal modeling language and the ARIS framework. As

an outcome of the semantic analysis, we were able to provide a correspondence

between subsets of these languages, in addition to clarifying the semantics of the main

goal-related constructs of these languages. We have concluded that the relations

between goals and between the goal domain and the business process domains as currently addressed in the ARIS method are overly simplistic and have opted to

employ the Tropos concepts to address this deficiency. The use of the ontology has

influenced heavily the definition of correspondences between the elements of both

approaches and also has revealed a significant difference in the notions of Objective

(ARIS) and Goal (Tropos), and their relations with respect to Functions (ARIS) and

Agents (Tropos). Further, we have been able to provide an initial account for the

notions of hard- and softgoals as well as the for relations between goals.

As future work, we intend to analyze the ARMOR language which extends

ArchiMate enterprise architecture modeling language with i*/Tropos concepts [28]. In

addition, we intend to enrich the semantic foundation with other goal relations to

provide a precise account for the notion of goal conflicts. Moreover we will address

the relations between goals and other elements of an enterprise architecture which are not currently covered in Tropos (nor ARMOR). In an exploratory case study that we

have conducted at a Hospital (in which Tropos and ARIS have been used,

respectively, for goal and enterprise modeling), we have concluded that the goal

domain has complex relations with a number of domains in Enterprise Modeling

beyond the business process and resource aspects as addressed in Tropos (and

reported in [6]). For example, there are relations between goals and organizational

norms (business rules, business policies as identified in the Business Motivation

Metamodel (BMM)), goals and the organizational structure, goals and agent’s skills

or capabilities, goals and agent’s beliefs, goals and properties of resources, among

others.



REFERENCES

1. Almeida, J. P. A., Guizzardi, G., Santos, P. S., Applying and Extending a Semantic Foundation for Role-Related Concepts in Enterprise Modelling, Enterprise Information Systems, v3, pp. 253-277 (2009)

2. Andersson, B., Bergholtz, M., Edirisuriya, A., Ilayperuma, T., Johannesson, P., Zdravkovic, J., Using Strategic Goal Analysis for Enhancing Value-based Business Models. Andersson, 2th International Workshop on Business/IT Alignment and Interoperability (BUSITAL’07).

Workshop at CAiSE’07 The 19th International Conference on Advanced Information Systems Engineering (2007)

3. Bottazzi, E. and Ferrario, R., Preliminaries to a DOLCE Ontology of Organizations, International Journal of Business Process Integration and Management, Vol. 1, pp.64–74 (2006)

4. Bresciani, P. and Sannicolo, F., Requirements Analysis in Tropos: a self referencing example. Technical Report DIT-02-105, University of Trento (2002)

5. Bresciani, P., Perini, A., Giorgini, P., Giunchiglia, F., Mylopoulos, J., Tropos: An Agent-

Oriented Software Development Methodology, Autonomous Agents and Multi-Agent Systems (2004)

6. Cardoso, E.C.S., On the Alignment between Goals Models and Enterprise Models with an Ontological Account, Master Thesis, Federal University of Espírito Santo, Brazil (2009) (http://nemo.inf.ufes.br/ecardoso)

7. Chung, L., Nixon, B., Yu, E. and Mylopoulos, J., Non-functional requirements in software engineering. Kluwer Academic Publishers (2000)

8. Davenport, T., Process Innovation: Reengineering work through Information Technology.

Harvard Business School (1992) 9. Davis, R., Brabander, E., ARIS Design Platform – Advanced Process Modelling and

Administration, Springer-Verlag (2008) 10. Davis, R., Business Process Modeling with ARIS - A Pratical Guide, Springer (2001) 11. Eclipse Foundation “Eclipse Modeling - EMF - Home”,

http://www.eclipse.org/modeling/emf/ (2008) 12. Giorgini, P., Mylopoulos, J., Perini, A., Susi, A., The Tropos Metamodel and its Use,

Informatical journal (2005)

13. Guizzardi, G. and Wagner, G., On A Unified Foundational Ontology and some Applications of it in Business Modeling, In 16th International Conference on Advances in Information Systems Engineering (Latvia), CAiSE 2004, Open INTEROP Workshop on Enterprise Modelling and Ontologies for Interoperability (2004)

14. Guizzardi, G., Falbo, R. A., and Guizzardi, R. S. S., Grounding Software Domain Ontologies in the Unified Foundational Ontology (UFO): The case of the ODE Software Process Ontology, In XI Iberoamerican Workshop on Requirements Engineering and Software Environments (IDEAS’2008), Brazil (2008)

15. Guizzardi, G., Ferreira Pires, L., and Van Sinderen, M., An Ontology-Based Approach for

Evaluating the Domain Appropriateness and Comprehensibility Appropriateness of Modeling Languages, In ACM/IEEE 8th International Conference on Model Driven Engineering Languages and Systems, Lecture Notes in Computer Science LNCS 3713. Springer-Verlag (2005)

16. Guizzardi, G., Ontological Foundations for Structural Conceptual Models, Ph.D. Thesis, CTIT PhD-thesis, University of Twente, The Netherlands (2005)

17. Guizzardi, R,, Guizzardi, G., Perini, A., Mylopoulos, J., Towards an Ontological Account of Agent Oriented Goals. Software Engineering for Multi-Agent Systems, Vol. 5, Springer-

Verlag (2007)

http://dev.nemo.inf.ufes.br:8280/files/applying_and_extending_a_semantic_foundation_for_role_related_concepts_in_enterprise_modelling_2009.pdf



http://dev.nemo.inf.ufes.br:8280/ecardoso

http://www.eclipse.org/modeling/emf/

http://www.inf.ufes.br/~gguizzardi/IDEAS2008Proceedings.pdf





18. Guizzardi, R., Agent-oriented Constructivist Knowledge Management. PhD Thesis, Twente University, The Netherlands (2006)

19. Halleux, P., Mathieu, L., Andersson, B., A Method to Support the Alignment of Business Models and Goal Models, In Proc. 3rd Workshop on Business/IT-Alignment and

Interoperability (BUSITAL'08), CEUR Workshop Proceedings (2008) 20. Hammer, M. and Champy, J., Reengineering the Corporation: A Manifesto for Business

Revolution, Nicholas Brealey Publishing, England (1993) 21. Hammer, M., Reengineering Work: Don’t Automate, Obliterate, Harvard Business Review

(1990) 22. Harrington, H. J., Business Process Improvement: The Breakthrough Strategy for Total

Quality, Productivity and Competetiveness. MacGraw-Hill, New York (1991) 23. Jonkers, H. et al, Concepts for Modelling Enterprise Architectures. International Journal of

Cooperative Information Systems (2004)

24. Jonkers, H., et al, Enterprise architecture: Management tool and blueprint for the organization, Information Systems Frontiers, Vol. 8 (2006)

25. Jureta, I. J., Faulkner, S., Schobbens P. Y., A More Expressive Softgoal Conceptualization for Quality Requirements Analysis, In Conceptual Modeling - ER, 25th International Conference on Conceptual Modeling, Arizona (2006)

26. Letier, E., van Laamsweerde, A Reasoning about Partial Goal Satisfaction for Requirements and Design Engineering, ACM SIGSOFT Software Engineering Notes. Volum 29, Issue 6 (2004)

27. Mylopoulos, J., Chung, L., Yu, E., Nixon, B., Representing and Using Non-functional Requirements: A Process-Oriented Approach, IEEE Trans. on Software Engineering, 18(6), pp. 483- 497 (1992)

28. Quartel, D., Engelsman, W. and Jonkers, H., A Goal-Oriented Requirements Modelling Language for Enterprise Architecture. Proceedings of the 13th IEEE International Enterprise Distributed Object Computing Conference, EDOC 2009, New Zealand (2009)

29. Rockart J., Chief executives define their own data needs, Harvard Business Review, 57(2), pp 81-93 http://www.ncbi.nlm.nih.gov/pubmed/10297607 (1979)

30. Santos Jr., P. S., Almeida, J. P. A., and Pianissolla, T. L., Uncovering the Organizational Modelling and Business Process Modelling Languages in the ARIS Method, International Journal of Business Process Integration and Management (IJBPIM), in press

31. Santos, P. S., Almeida, J. P. A., Guizzardi, G., An Ontology-Based Semantic Foundation for ARIS EPCs, In: 25th ACM Symposium on Applied Computing (Enterprise Engineering Track), Switzerland (2009)

32. Scheer, A.W., ARIS – Business Process Modeling. Third Edition. Springer (1999) 33. Sharp, A. and McDermott, P., Workflow Modeling-Tools for Process Improvement and

Application Development. Artech House (2001) 34. Sowa, F., Zachman, J., Extending and formalizing the framework for information systems

architecture, IBM Systems Journal, v.31, p 590-616 (1992) 35. Yu, E., Modelling Strategic Relationships for Process Reengineering, PhD Thesis,

University of Toronto (1995) 36. Yu, E., Strohmaier, M. and Deng, X., Exploring Intentional Modeling and Analysis for

Enterprise Architecture, In Proceedings of the EDOC 2006 Conference Workshop on Trends in Enterprise Architecture Research (TEAR 2006). IEEE Computer Society, Hong

Kong (2006) 37. Zachman, J., A Framework for Information Systems Architecture, IBM Systems Journal.

pp. 276-292 (1987)

http://doc.utwente.nl/56967/1/r_guizzardi.pdf

Date post:	20-Jan-2022
Category:	Documents
Upload:	others
View:	2 times
Download:	0 times

Semantic Integration of Goal and Business Process Modeling

Documents